Skip to content

Conversation

@huydhn
Copy link
Contributor

@huydhn huydhn commented Dec 13, 2025

Several small tweaks:

  • Disable arm benchmark for now as it's not working and there is no owner
  • Only need to parse the list of model from serving benchmark config. It always has all the models because serving benchmark is the most basic one, and it has tensor_parallel_size field set correctly. This is to fix a non deterministic issue where a model's tensor_parallel_size could be missing when glob decides to return the list of benchmark configs differently
  • Run daily, saving the capacity for the upcoming nightly run

@huydhn huydhn requested a review from yangw-dev December 13, 2025 08:32
@meta-cla meta-cla bot added the cla signed label Dec 13, 2025
Signed-off-by: Huy Do <huydhn@gmail.com>
@huydhn huydhn force-pushed the vllm-benchmark-tweaks-1213 branch from 043f5b1 to c4dcf9f Compare December 13, 2025 08:34
Signed-off-by: Huy Do <huydhn@gmail.com>
Signed-off-by: Huy Do <huydhn@gmail.com>
@huydhn huydhn merged commit c4ccc4d into main Dec 15, 2025
1 check passed
fadara01 added a commit that referenced this pull request Dec 16, 2025
- re-enables the benchmarks disabled in #114
- related to vllm-project/vllm#26494 (not sure which needs to go in first)

Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
fadara01 added a commit that referenced this pull request Dec 16, 2025
- re-enables the benchmarks disabled in #114
- related to vllm-project/vllm#26494 (not sure which needs to go in first)
- use default block_size in serving benchmarks (i.e. 128 instead of setting it to 16)

Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
fadara01 added a commit that referenced this pull request Dec 17, 2025
- re-enables the benchmarks disabled in #114
- related to vllm-project/vllm#26494 (not sure which needs to go in first)
- use default block_size in serving benchmarks (i.e. 128 instead of setting it to 16)

Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
fadara01 added a commit that referenced this pull request Dec 17, 2025
- re-enables the benchmarks disabled in #114
- related to vllm-project/vllm#26494 (not sure which needs to go in first)
- use default block_size in serving benchmarks (i.e. 128 instead of setting it to 16)

Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants